Multioutput support vector regression for remote sensing biophysical parameter estimation Articles uri icon

authors

  • TUIA, DEVIS
  • VERRELST, JOCHEM
  • ALONSO, LUIS
  • PEREZ CRUZ, FERNANDO
  • CAMPS-VALLS, GUSTAVO

publication date

  • July 2011

start page

  • 804

end page

  • 808

issue

  • 4

volume

  • 8

International Standard Serial Number (ISSN)

  • 1545-598X

Electronic International Standard Serial Number (EISSN)

  • 1558-0571

abstract

  • This letter proposes a multioutput support vector regression (M-SVR) method for the simultaneous estimation of different biophysical parameters from remote sensing images. General retrieval problems require multioutput (and potentially nonlinear) regression methods. M-SVR extends the single-output SVR to multiple outputs maintaining the advantages of a sparse and compact solution by using an e-insensitive cost function. The proposed M-SVR is evaluated in the estimation of chlorophyll content, leaf area index and fractional vegetation cover from a hyperspectral compact high-resolution imaging spectrometer images. The achieved improvement with respect to the single-output regression approach suggests that M-SVR can be considered a convenient alternative for nonparametric biophysical parameter estimation and model inversion.This letter proposes a multioutput support vector regression (M-SVR) method for the simultaneous estimation of different biophysical parameters from remote sensing images. General retrieval problems require multioutput (and potentially nonlinear) regression methods. M-SVR extends the single-output SVR to multiple outputs maintaining the advantages of a sparse and compact solution by using an e-insensitive cost function. The proposed M-SVR is evaluated in the estimation of chlorophyll content, leaf area index and fractional vegetation cover from a hyperspectral compact high-resolution imaging spectrometer images. The achieved improvement with respect to the single-output regression approach suggests that M-SVR can be considered a convenient alternative for nonparametric biophysical parameter estimation and model inversion.

keywords

  • geophysical image processing; image resolution; regression analysis; remote sensing; spectrometers; support vector machines; vegetation mapping